Association of decreased natural and antibody-dependent cellular cytotoxicity and production of natural killer cytotoxic factor and interferon in neonates

CELLULAR

IMMUNOLOGY

94, 159-l 7 1 (1985)

Association of Decreased Natural and Antibody-Dependent Cytotoxicity and Production of Natural Killer Cytotoxic and Interferon in Neonates’

Cellular Factor

MADHAVAN P. N. NAIR, **' STANLEY A. SCHWARTZ,*,~ ANDMANGALADEVI MENoN? Departments

of *Pediatrics University Received

and Epidemiology and j-obstetrics and Gynecology, Ann Arbor, Michigan 48109

of Michigan,

February

4, 1985; Accepted

March

15, I985

Cord blood lymphocytes (CBL) were compared with adult peripheral blood lymphocytes (a-PBL) for their: (i) natural killer (NK) and antibody-dependent cellular cytotoxic (ADCC) activities, (ii) target-binding capacity, (iii) ability to induce soluble natural killer cytotoxic factor (NKCF), (iv) interferon (IFN>, interleukin 2 (IL-2)-, and lectin-induced augmentation of NK activity, and (v) ability to produce IFN against tumor targets in vitro. CBL depleted of adherent cells and Percoll-separated, NK-enriched subpopulations demonstrated significantly lower NK, ADCC, and target-binding activities compared to a-PBL. CBL produced significantly lower levels of NKCF directed against K562 tumor targets in comparison with a-PBL. Although the NK activity of CBL was not stimulated by either IFN or IL2 to the same levels shown by a-PBL, the percentage enhancement of cytotoxicity of CBL by IFN and IL-2 was greater than that of a-PBL. Lectin-induced enhancement of cytotoxicity was significantly greater for CBL in comparison with a-PBL. Further, the ability of CBL lymphocytes to produce EN--r in vitro against K562 target cells was significantly lower than that of adult PBL. These studies suggest an association between decreased NK, ADCC, and target-binding activities, induction of NKCF and IFN production by CBL, and increased susceptibility of neonates to infection. o 1985 Academic

F’ress, Inc.

INTRODUCTION Neonates are known to have increased susceptibility to disseminated and devastating infections (1, 2). Earlier studies showed that the immune system of the newborn is deficient in several aspects of humoral and cellular immunity (1, 3). Natural killer (NQ4 cells and cells mediating antibody-dependent cellular cytotoxic ’ This work was aided by grants from the National Institutes of Health (CA35922 and AI-19890), the Children’s Leukemia Foundation of Michigan, and the Michigan Diabetes Research and Training Center of the University of Michigan. ‘To whom all correspondence should be addressed at: Room 1029, School of Public Health I, Department of Pediatrics, University of Michigan, Ann Arbor, Michigan 48 109. 3 Recipient of NIH Research Career Development Award CA-08896. 4 Abbreviations used: ADCC, antibody-dependent cellular cytotoxicity; a-PBL, adult peripheral blood lymphocytes depleted of adherent cells; CBL, cord blood lymphocytes depleted of adherent cells; NK, natural killer; NKCF, natural killer cytotoxic factor; EN, interferon; IL-2, interleukin 2; LGL, large granular lymphocytes; FCS, fetal calf serum; LDCC, lectindependent cell-mediated cytotoxicity; PHA, phytohemagglutinin; LU, lytic units; Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid. 159 0008-8749185 $3.00 Copyright 0 1985 by Academic Press, Inc. All rights of reproduction in any form reserved.

160

NAIR,

SCHWARTZ,

AND

MENON

(ADCC) activities are important because of their probable role in immune surveillance against cancer and viral infections and in the regulation of various lymphocyte reactions (4-7). Regulatory effects of NK cells on hematopoietic stem cell proliferation have been demonstrated (8). The ability of cord blood lymphocytes (CBL) to mediate NK and ADCC activities has been reported with conflicting findings (914). The present study examines the NK and ADCC activities of CBL in comparison with adult lymphocytes (a-PBL) and attempts to delineate the molecular and cellular basis of the NK defect by examining the target-binding ability and production of natural killer cytotoxic factor (NKCF) and interferon (IFN) in vitro. These studies are consistent with a relationship between the deficiency of NK, ADCC, and targetbinding activities of cord blood lymphocytes and reduced NKCF and IFN production. MATERIALS Preparation

AND

METHODS

of Efector Cells

Neonatal umbilical cord blood samples were collected in heparinized tubes (20 U/ml) under sterile conditions from healthy term infants at vaginal delivery before the expulsion of the placenta. Heparinized (20 U/ml) peripheral blood was obtained simultaneously from healthy unrelated adult volunteers of either sex (20 to 40 years of age). Blood samples were processed immediately after collection. Donors or their parents were apprised of the study and consents were obtained consistent with the policies of the University of Michigan and the NIH. Mononuclear cells from both adult and cord blood samples were isolated using a modified method of Boyum (15). Blood was diluted with an equal volume of Ca2’- and Mg2+-free Hanks’ balanced salt solution [HBSS (GIBCO, Grand Island, N.Y.)] and centrifuged at 400g for 30 min at 18°C over a cushion of Ficoll-Hypaque (Pharmacia Fine Chemicals, Piscataway, N.J.). The mononuclear cell band was harvested, washed three times with HBSS, and resuspended in RPM1 1640 medium containing 25 mM Hepes buffer supplemented with 10% heat-inactivated fetal calf serum [FCS (GIBCO)], 80 &ml gentamicin (Schering Corp., Kenilworth, N.J.), and 300 pg fresh glutamine/ml (complete medium). Depletion of Adherent Cells Mononuclear cells were depleted of adherent cells as described (6). Briefly, total mononuclear cells were suspended in RPM1 1640 with 10% FCS and passaged through a 7-ml column of Sephadex G-10 beads (Pharmacia) equilibrated in the same medium. After 45 min of incubation at 37°C nonadherent cells were washed through with 1 bed vol of warm (37°C) medium. The cell recovery was >70% of the total input, and monocyte contamination as indicated by nonspecific esterase staining was <2%. Percoll Fractionation

of Lymphocytes

The enrichment of NK and ADCC effector cells using a discontinuous gradient of Percoll (Pharmacia) was carried out as described elsewhere (6). To prepare a discontinuous density gradient, Percoll solution was mixed at various concentrations with RPM1 1640 medium and 2-ml aliquots (ranging from 50.0 to 37.5% in 2.5% increments) were gently layered into 15 X 130-mm round-bottom glass test tubes.

ACTIVITY

OF CORD BLOOD

LYMPHOCYTES

161

Adherent cell-depleted lymphocytes were layered on top of the gradient and centrifuged at 300g for 45 min at 20°C. Six fractions, with the uppermost (37.5%) designated “0” and the lowermost (50.0%) designated “V”, were collected from the top with a Pasteur pipet and washed twice in RPM1 1640 medium. Recovery of cells was -85% of the input, and viability always exceeded 95% by trypan blue dye exclusion. Preparation

of Tumor

Target Cells

Mycopiasma-free human erythroleukemia cell line K-562 was used as target for NK cells. For the ADCC assay, we chose as target an antibody-coated human B-cell leukemia line (SB) because of its resistance to NK activity in a 4-hr “Cr release assay. Tumor target cells were serially passaged in complete medium and used in cytotoxicity assays no more than 48 hr after the last passage. To 0.8-ml aliquots of complete medium containing 5 X lo6 washed tumor cells, 200 PCi of 51Cr as sodium chromate (New England Nuclear, Boston, Mass.) was added. The cells were incubated at 37°C for 1 hr in a humidified atmosphere of 5% CO* in air, with intermittent shaking. After incubation, the cells were washed three times with complete medium and resuspended to a concentration of 1 X lo5 cells/ml. Assay for NK Activity NK activity was measured in a direct “Cr release assay as previously described (6, 16, 17). A fixed number of viable effector cells in complete medium was added to triplicate cultures of “Cr-labeled targets in V-bottom microtitration plates (CoStar, Cambridge, Mass.). After centrifugation at 40g for 2 min, they were incubated at 37°C in a humidified atmosphere of 5% COz in air for 4 hr. At the end of incubation the plates were centrifuged at 400g for 10 min and 100-d aliquots were removed from each well and transferred to glass tubes and read in a Packard Model 593 gamma counter. Percentage cytotoxicity was calculated as follows: % cytotoxicity

=

experimental release - spontaneous release x 100, total release - spontaneous release

where spontaneous release represents counts released from control wells containing only 1 X lo4 target cells, and total release represents counts obtained from an aliquot of 1 X lo4 target cells. Cytotoxicity was also expressed as lytic units (LU)/ 10’ effector cells, defined as the number of effector cells needed to yield 30% cytotoxicity of 1 X IO4 target cells. LU were calculated from the cytotoxicity curve using four different E:T cell ratios for each test by linear regression analysis as described by Kadish et al. (18). Assay for ADCC The ADCC activity of effector cells was determined as previously described (6, 16, 17). Briefly, 50 ~1 of varying concentrations of effector cells were added to 50 ~1 of complete medium containing 51Cr-labeled SB target cells and 100 ~1 of a 2 X 10e4 dilution of rabbit anti-SB antisera previously found to yield maximum ADCC. Percentage cytotoxicity was calculated as described above for NK activity with the following exceptions. Spontaneous release represents counts released in

NAIR, SCHWARTZ,

162

AND MENON

control wells containing effector cells, labeled SB target cells, and media instead of anti-SB antibodies, and total release represents counts obtained in an aliquot of 1 X lo4 51Cr-labeled target cells. Lectin-Dependent

Cell-Mediated

Cytotoxicity

(LDCC) Assay

The LDCC assay was carried out in a manner similar to that of the NK assay except that NK-resistant SB cells were used as targets. Briefly, different numbers of target and effector cells were added to V-bottom microtitration plates and various concentrations of phytohemagglutininP [PHA, Difco, Detroit, Mich. (1, 2.5, and 5 Fg/ml)] were added to a final volume of 0.2 ml and incubated for 4 hr at 37°C in a humidified 5% CO2 in air incubator. Percentage LDCC activity was calculated as described for NK activity and was compared with cytotoxicity observed with control cultures containing effector and target cells in the absence of PHA. Percentage cytotoxicity was calculated as % cytotoxicity

=

experimental Release - spontaneous Release x 100, total Release - spontaneous Release

where spontaneous release represents counts released from control wells containing 1 X lo4 target cells plus lectin and total release represents counts obtained in an aliquot of 1 X lo4 target cells. The cytotoxicity was expressed as lytic units (LU)/ lo7 effector cells as described for NK assay. Interferon (IFN) Recombinant IFN-(I! (specific activity, - 1 X 10’ U/mg protein) was a gift from the Schering Corporation, Bloomfield, New Jersey. IFN-(u was diluted in RPM1 1640 and stored at -70°C before use. The maximum NK-augmenting dose was 250 U/ml as previously determined. IFN-a was diluted in RPM1 1640 medium and stored at -70” before use. Interleukin

2 (IL-2)

Commercially available IL-2 (Electra-Nucleonics Labs, Inc., Silver Spring, Md.) was diluted in RPM1 1640 and stored at -70°C before use. IL-2 was purified from human peripheral blood leukocytes by several chromatographic steps and was devoid of any lectin or IFN. A 10% final concentration, the maximum NKaugmenting dose as previously determined, was used. Treatment

of E_ff^ectorCells with IFN or IL-2

Effector cells (2 X 106/ml) were washed and suspended in 1 ml of RPM1 1640 plus 5% FCS to which IFN-(I! or IL-2 was added. Cultures were incubated for 24 hr at 37°C in a humidified atmosphere of 5% COz in air, washed twice, and resuspended in medium. Control cultures were treated identically with the exception that either IFN-ar or IL-2 was not added. Viability of treated lymphocytes was unaffected as assessed by trypan blue dye exclusion. Both treated and control cultures were assayed for NK activity.

ACTIVITY

Target-Binding

OF CORD BLOOD

LYMPHOCYTES

163

Assay

A target-binding assay was used as described by Haliotis et al. (19). In brief, lymphocytes were washed twice and mixed with a fivefold excess of tumor cells, centrifuged at 15Og for 5 min, and incubated at 37°C for 5 min. Following incubation, lymphocyte-target mixtures were transferred to ice, the pellets were gently resuspended, and the number of lymphocytes binding to tumor targets was determined using a hemocytometer. Interferon Assay Lymphocytes were cultured alone or with K562 target cells at a ratio of 1OO:l for 18 hr at 37°C in 5% CO1 in air, after which they were centrifuged and the supemates were collected for IFN assay. Supemates were quantitated for IFN using a modified calorimetric assay of Borden and Leonhardt (20). Briefly, 96-well flatbottom microtitration plates (Costar) were seeded with the human epithelial cell line WISH at a concentration of 3 X lo4 cells/well in 0.1 ml of media and incubated for 18 hr. Serial dilutions of supernates (1: 10, 1:40, 1: 160, and 1:320) were added in 0.1 ml and incubated again for I8 hr. The cells were then infected with vesicular stomatitis virus, 200 plaque forming units/well, for 24 hr. After development of cytopathic effect the cultures were washed with PBS and overlaid with 0.2 ml of 0.015% neutral red in saline and incubated for 2 hr at 37°C. The monolayers were washed and the cell-bound dye was eluted with a 1:l solution of absolute ethanol: 0.2 M NaHzP04. As a control, an IFN preparation standardized against NIH reference GO23-901-527 of known concentrations ranging from 2 to 300 U/ml was included in the assay. This assay is highly reproducible and can detect as little as 4 U/ml of IFN. Generation of Soluble Natural

Killer Cytotoxic Factor (NKCF)

The preparation of NKCF was carried out as described by Farram and Targan (21). Briefly, effector cells were mixed with NK-sensitive unlabeled K562 target cells at a 1OO:l effector-to-target-cell ratio for 24 hr in sterile V-bottom microtitration plates (Dynatech Labs, Alexandria, Va.). They were then centrifuged at 15Og for 10 min, and the supemates (NKCF) were collected and tested for cytotoxic activity against fresh target cells, As controls, either K562 or effector cells were incubated separately in media alone and supemates were collected as described above and used as mock NKCF. Assay for NKCF K562 target cells were labeled with 5’Cr as described above and suspended at a concentration of 1 X lo5 cells/ml to which NKCF or mock NKCF at a final concentration of 25% (v/v) was added and incubated for 24 hr at 37°C in 5% CO2 in air. At the end of incubation, the plates were centrifuged and loo-p1 aliquots were removed from each well and counted as described for the NK assay. Percentage cytotoxicity was calculated as % cytotoxicity = experimental release (NKCF) - spontaneous release (Mock NKCF) total release - spontaneous release (Mock NKCF)

x loo

164

NAIR, SCHWARTZ,

AND MENON

RESULTS NK and ADCC Activities of Cord Blood Lymphocytes Lymphocytes depleted of adherent cells from 15 cord blood and 10 adult peripheral blood samples were tested for their NK and ADCC activity against tumor target cells. Data presented in Table 1 show that CBL depleted of adherent cells demonstrated significantly lower NK and ADCC activities, respectively, against K562 and antibody-coated SB target cells compared to normal adult lymphocytes at all E:T cell ratios. E:T cell ratios higher than 5O:l produced similar differences (data not shown). The possibility that dilution of NK cells with other cells of the neonatal lymphocyte pool may be responsible for the diminution of NK and ADCC activities was examined using Percoll-enriched NK and ADCC effector cells. The data presented in Fig. 1 reveal that NK- and ADCC-enriched CBL from fraction “0” also showed significantly diminished NK (25%, P < 0.025) and ADCC (21%, P < 0.01) activities compared to higher NK (68%) and ADCC (52%) activities displayed by a similar Percoll fraction of a-PBL. CBL from fraction II also manifested lower levels of NK (8%, P < 0.05) and ADCC activities (12%, P < 0.025) as contrasted with the levels of NK (26%) and ADCC (18%) activities manifested by the same fraction of a-PBL. Lymphocytes from nonlytic fractions II to V of both CBL and a-PBL did not manifest significant differences in NK or ADCC activities. Target-Binding

Ability

of CBL

To examine whether the diminished cytotoxicity observed with CBL is due to failure to interact with a target recognition structure rather than inhibition of the lytic mechanism, target-binding assays were performed. The results presented in TABLE 1 NK and ADCC Activities of Cord and Adult Peripheral Blood Lymphocytes” Effector:target cell ratio Cytotoxicity (%)b Targets

Effector source

K562’

a-PBL’

5o:l

25:l

1O:l

51.2 + 15.6

19.6 + 5.0

12.6 f 2.3

(P < 0.025)

(P < 0.005)

13.1 + 6.1

31.1 f 10.0 (P < 0.01) 9.2 + 4.6

6.2 f 3.3

3.6 + 2.1

39.5 *

29.4 *

(P < 0.025)s CBI! SBd

a-PBL CBL

23.6 + 5.6

8.5 f 1.9

(P < 0.05)

(P < 0.05)

(P < 0.025)

(P < 0.025)

12.4 zk 9.6

10.5 + 6.3

1.7 + 3.2

2.6 k 1.7

9.5

8.2

5:l

a Effector cells were mixed with targets at varying E:T cell ratios in a 4-hr 5’Cr release assay.The results are means + SD of 15 different cord blood lymphocyte samples and 10 unrelated adult peripheral blood lymphocyte samples. b Percentage cytotoxicity was calculated as under Materials and Methods. ’ Erythroleukemia cell line used as targets in the NK assay. d B-Leukemia cell line sensitized with anti-SB antibody used as targets in the ADCC assay. ’ Adult peripheral blood lymphocytes depleted of adherent cells. ‘Cord blood lymphocytes depleted of adherent cells. 8 Differences in values between CBL and a-PBL were statistically significant at all E:T ratios in both NK and ADCC assaysusing a single-tailed Student’s t test.

ACTIVITY

OF CORD BLOOD l

I

1

LYMPHOCYTES

165

NK-a-PBL 0 NK-CBL

FIG. 1. NK and ADCC activities of PercoB-enriched cord blood lymphocytes and adult peripheral blood lymphocytes. Values represent mean percentage cytotoxicity + SD of three separate experiments with triplicate determinations for each experiment at 20~1 E:T cell ratio. Statistical significance was determined using a single-tailed Student’s t test.

Table 2 show that CBL demonstrated a lower percentage of conjugates formed with target cells (6.5%) compared to a-PBL (13.5, P < 0.05). This suggests that the NK deficiency of CBL may be associated with defects in binding events.

Efect of Lectin on Cellular Cytotoxicity of CBL Since CBL demonstrated reduced NK and ADCC activities and target binding ability compared to a-PBL, we examined whether the cytotoxic activity of CBL could be stimulated with lectin. Data presented in Table 3 demonstrate that a-PBL produced significantly increased cytotoxicity, i.e., 19 (P < 0.05) 22 (P < 0.05) and 18 (P < 0.025) LU against SB targets at 1, 2.5, and 5 pg/ml of PHA, respectively, compared to a negligible cytotoxicity (5 LU) displayed in the absence of PHA. CBL exhibited a cytotoxicity of 4 LU without PHA which was significantly enhanced to 31 (P < 0.05), 31 (P < 0.05), and 49 (P < 0.05) LU by PHA at 1, 2.5, and 5 pg/ TABLE 2 Target Binding Ability of Cord Blood Lymphocytes” Effector cells

Target-binding lymphocytes @o)~

a-PBL’

13.5 * 3.1 (P < 0.05)’ 6.5 + 2.4

CBL*

a Effector and target cells were mixed at an ET ratio of 5: 1 and incubated at 37°C for 30 min, followed by 0°C for 1 hr. Aliquots were examined in a hemocytometer and % positive target-binding cells was scored for each lymphocyte binding two or more target cells. Results are expressed as means + SD of 15 CBL and 7 a-PBL samples performed in duplicate. b The human erythroleukemia cell line, K562, was used as target in the binding study. ’ Adult peripheral blood lymphocytes depleted of adherent cells. *Cord blood lymphocytes depleted of adherent cells. e Statistical significance was determined using a single-tailed Student’s t test.

166

NAIR, SCHWARTZ,

AND MENON

ml concentrations, respectively. These data demonstrate that the percentage enhancement of cytotoxicity by PHA was greater for CBL than for a-PBL. Ability of CBL to Produce Soluble NKCF We further investigated whether a decreased ability of CBL to synthesize and release NKCF might be responsible for their diminished cytotoxicity when compared to a-PBL. Data presented in Table 4 show that culture supernates (NKCF) of CBL plus target cells manifested decreased cytotoxicity ( 14%) against fresh K562 target cells as compared to a higher cytotoxicity (24%) displayed by NKCF produced by a-PBL (P < 0.05). This suggests that diminished production and secretion of NKCF may, in part, explain the decreased cytotoxicity of CBL. Efect of IFN-a

and IL-2 on the NK Activity of CBL

IFN-(I! and IL-2, agents known to augment the NK activity of lymphocytes, were examined for their potential to enhance the decreased cytotoxicity of CBL (Table 5). We observed that a-PBL precultured with IFN-a for 24 hr produced higher cytotoxicity (137 LU) compared to a lower cytotoxicity (86 LU) displayed by untreated lymphocytes (P < 0.05). Preculture of a-PBL with IL-2 also demonstrated a significantly higher cytotoxicity of 173 LU compared to 86 LU shown by untreated a-PBL (P < 0.0 1). CBL precultured with IFN and IL-2 also showed an enhancement of their cytotoxicity, 122 LU (P < 0.05) and 183 LU (P < O.OOS), respectively, compared to 58 LU produced by untreated CBL. These data indicate that although the cytotoxicity of CBL could not be enhanced to absolute values comparable to a-PBL levels, the percentage enhancement of cytotoxicity of CBL by IFN-a (110.6%) and IL-2 (215%) was remarkably higher than a-PBL augmented by IFN (59%) and IL-2 ( 110%). TABLE 3 Effect of Lectin on Cellular Cytotoxicity of Lymphocytes’ Cytotoxicity (LU)* Lymphocyte source

PHA concentrations 0

1

2.5

5

a-PBL’

5.4 + 2.9

CBLd

4.6 + 3.2

19.8 +- 5.7 [266]’ (P < 0.05)J 31.2 + 11.7 [578] (P < 0.05)

22.8 f 6.4 [322] (P < 0.025) 31.3 + 11.0 [580] (P < 0.05)

18.7 + 4.6 [246] (P < 0.025) 49.7 + 18.5 [980] (P < 0.05)

(&ml):

p The LDCC assaywas carried out in a manner similar to that of the NK assay except the NK-resistant SB cell line was used as targets as described under Materials and Methods. * Cytotoxicity is expressed as LU/ 10’ effector cells as described under Materials and Methods. Values are mean LU + SD of four separate experiments performed in triplicate. c Adult peripheral blood lymphocytes depleted of adherent cells. d Cord blood lymphocytes depleted of adherent cells. ’ Values in brackets are % enhancement of LDCC caIcu1ated in comparison to control. ‘Statistical values were determined by a single-tailed Student’s t test.

ACTIVITY

OF CORD BLOOD

167

LYMPHOCYTES

TABLE 4 Effect of Soluble Natural Killer Cytotoxic Factor from Cord and Adult Peripheral Blood Lymphocytes on K562 Targets” Source of NKCF

Cytotoxicity ’ (%)

a-PBL + K562’

24.2 f 4.1 (P < 0.05)’ 13.8 + 1.9

CBL + K562*

n Adult PBL or CBL were precultured separately with NK-sensitive unlabeled K562 target cells at an E:T cell ratio of lOO:l, for 24 br at 37°C in 5% CO2 in air. The supemates were collected and used as putative NKCF against fresh 5’Cr-lab&d K562 target cells at a final concentration of 25% (v/v) as described under Materials and Methods. Results are the means of 10 different CBL-NKCF and 7 different a-PBL-NKCF samples + SD performed in triplicate. * Calculated as under Materials and Methods. c Adult peripheral blood lymphocytes depleted of adherent cells + unlabeled K562 target cells. *Cord blood lymphocytes depleted of adherent cells + unlabeled K562 target cells. e Differences in values between PBL and CBL were statistically significant using a single-tailed Student’s I test.

Interferon-y

Production

by CBL in Vitro

Susceptibility of target cells to NK lysis may be partially dependent upon the ability of effector lymphocytes to produce IFN against the target cells (22). As CBL demonstrated diminished NK activity, target binding, and NKCF production compared to a-PBL, we examined whether the ability of CBL to produce IFN-?/ in vitro against K562 tumor targets could also be correlated with their reduced cytotoxic response. Data presented in Table 6 demonstrate that all CBL samples produced

TABLE 5 Effect of IFN and IL-2 on NK Activity of Lymphocytes” Effector cells

Treatment

Cytotoxicity (LU)b

Percentage enhancement’

a-PBL* a-PBL a-PBL

None IFN IL-2

86.2 k 12.2 137.2 f 22.4 173.0 f 23.9

(P < 0.05)’ (P -=c0.01)

59.1 100.6

CBL’ CBL CBL

None IFN IL-2

58.2 + 8.2 122.6 + 21.6 183.5 f 32.3

(P c: 0.05) (P < 0.005)

110.6 215.2

’ Effector cells were cultured alone or with IFN (250 U/ml) or IL-2 (10% v/v) for 24 hr at 37°C washed, and tested for NK activity against prelabeled K562b target cells. b NK reactivity was calculated as LU/lO’ effector cells as under Materials and Methods. Values represent mean LU k SD of five separate experiments done in triplicate. ‘Percentage enhancement was calculated in comparison to the NK reactivity of untreated control lymphocytes. * Adult peripheral blood lymphocytes depleted of adherent cells. ’ Cord blood lymphocytes depleted of adherent cells. ‘Statistical significance was determined using a single-tailed Student’s t test.

168

NAIR,

SCHWARTZ,

AND

MENON

remarkably lower levels of IFN-7, with a range of 4 to 1680 U/ml (except for one sample, No. 5, which produced 3685 U/ml), in response to K562 target cells as compared to a higher level of IFN-7 produced by a-PBL. Thus, reduced IFN-7 production by CBL may also be associated, in part, with their decreased ability to mediate NK activity. DISCUSSION Since newborns are more susceptible to various infections than are adults (1, 2), analysis of their immune status is an important clinical parameter. Although deficiencies of both humoral and cell-mediated immunity have been observed in neonates, the underlying cellular and molecular mechanisms remain unclear. In humans, reports on the ontogeny of NK cells are conflicting (23-25). Defects in the ability of neonatal lymphocytes to mediate NK and ADCC activities against tumorand virus-infected targets have been described by several investigators (9, 11). Frazier et al. (26) showed that NK and ADCC activities of CBL against Herpes simplex virus (HSV)-infected Chang liver cells were lower than those of a-PBL. They further demonstrated that the NK and ADCC activities of lymphocytes from babies delivered by C-section without labor were significantly lower than the NK and ADCC activities of lymphocytes from neonates delivered vaginally. The proportion of certain lymphocyte subpopulations like OKTl l+ (27), OKT8+ (28), HNKl+ (12, 25) or Fc receptor+ (29) was also reported to be lower in newborns than adults. Numerous studies have shown the presence of active suppressor cells or their soluble mediators in CBL (27, 30). The present investigation (Table 1) shows that CBL from babies delivered vaginally demonstrate reduced NK and ADCC activities against tumor target cells compared to a-PBL. NK and ADCC effector cells have been described as large granular lymphocytes (LGL) containing azurophilic granules with a high cytoplasmic:nuclear ratio and a sedimentation profile at lower densities in a Percoll gradient (3 1). Tarkkanen and Saksela ( 14) showed an equivalent percentage of LGL in neonates and adult mononuclear populations. Although the percentage yield of LGL in Percoll fractions 0 and I are similar in both CBL and a-PBL populations (data not presented), the CBL subpopulation enriched on Percoll gradient fractions 0 and I demonstrated significantly lower NK and ADCC activities compared to an equal number of a-PBL similarly enriched (Fig. 1). CBL from fractions II to V (nonlytic fractions) showed negligible NK and ADCC activities, which suggested that CBL effector cells do not have sedimentation characteristics different from those observed with a-PBL. Further, we examined whether the diminished cytotoxicity of CBL may be due to a decreased target-binding ability of effector cells. It was observed that CBL exhibited a decreased target-binding capacity compared to a-PBL (Table 2), indicating that the reduced cytotoxicity of CBL may be partly due to a defective target-binding capacity in addition to a possible diminution of the intrinsic lytic mechanism of effector cells. This is consistent with the finding of Kohl et al. (32), who showed that total mononuclear cord blood cells manifest decreased early target cell recognition or binding against Chang liver target cells. Several investigators have observed that NK cells function through the release of soluble mediator(s), namely natural killer cytotoxic factor. Exogenous NKCF has been demonstrated to be cytotoxic to sensitive target cells (21, 33). In the present

ACTIVITY

OF CORD BLOOD

169

LYMPHOCYTES

report, we examined whether a decreased ability of CBL to synthesize and secrete NKCF is responsible for their reduced NK activity. Our results show that NKCF from CBL displayed significantly lower cytotoxicity compared to NKCF from a-PBL, suggesting that such a deficiency in NKCF production by CBL may be responsible, in part, for their decreased NK activity (Table 4). IFN and IL-2 play a major role in immunoregulation (6, 34-38) and the augmentation of NK activity both in vitro and in vivo (6, 34, 35). IFN appears to enhance NK activity by several mechanisms, including stimulation of noncytotoxic pre-NK cells to mature cytotoxic effector cells (24) or activation of mature NK cells that are transiently inactive (36, 37). Alternatively, IFN may potentiate endogenous lytic mechanisms by stimulating the development of new receptors or modulating preexisting effector cell surface molecules or their avidity (24, 36, 38). It has also been shown that IL-2 can induce IFN; this has been used successfully in the establishment of NK clones (39). Recently Svedersky et al. (40) reported that augmentation of NK activity by IL-2 and IFN involves overlapping mechanisms. The present investigation describes a higher relative response of CBL to IFN and IL-2 compared to a-PBL, although the cytotoxicity of the former was not enhanced to the same absolute levels shown by a-PBL. It is possible that a specific subpopulation of lymphocytes such as precursor effector or pre-NK cells, which may be more sensitive to IFN-(u and IL2 signals, may be more abundant in cord blood than in adult peripheral blood. Further, active cytotoxic effecters may be relatively decreased in cord blood compared to adult blood. Several investigators have demonstrated that spontaneous suppressor cell activity is greater in cord blood than in adult blood TABLE 6 Interferon Activity of Culture Supemates from Effector and Target Cells” Source of culture supemate

Interferon activity W/ml1 b

K562 K562 K562 K562 K562 K562 K562 K562 K562 K562

+ + + + + + + + + +

CBL-I CBL-2 CBL-3 CBL-4 CBL5 CBL-6 CBL-7 CBL-8 CBL-9 CBLlO

1680 700 230 <4 3685 420 40 160 240 60

K562 K562 K562 K562 K562

+ + + + +

a-PBL a-PBL a-PBL a-PBL a-PBL

2800 2677 2163 2334 2465

(1Mixtures of effector and unlabeled K562 target cells were set up at 1oO:l ET cell ratio at 37°C for 24 hr in 5% CO2 in air. Supemates were obtained from cell cultures and assayed for IFN as described under Materials and Methods. b Results are expressed as means + SD of triplicate determinations of 10 individual CBL and 5 different a-PBL samples.

170

NAIR, SCHWARTZ,

AND MENON

(3, 4 1), and cord blood cells are able to produce soluble suppressor factors which are capable of inhibiting various functions of adult lymphocytes (30). Enhancement of NK activity by IFN may also be due to the inhibition of suppressor cells or their soluble products. This is supported by our earlier investigation in which the suppressive effect of soluble suppressor factors produced by normal lymphocyte cultures on NK activity could be completely reversed by IFN (6). Bryson et al. (42) showed that human CBL can produce IFN-(u and IFN-~ but not IPN-y when stimulated with PHA. IPN has been shown to enhance the cytotoxicity of lymphocytes against tumor and virus-infected target cells. The magnitude of and sensitivity to IFN production in vivo may determine, in part, the neonate’s ability to resist infection. The present investigation demonstrates that CBL produce considerably less IFN-y in vitro in response to tumor target cells in comparison with a-PBL. Although our functional assay does not discriminate the type of IPN present, treatment of culture supernates containing IFN at acidic pH (pH 2-3) eliminated the IFN activity, demonstrating that IFN-y was produced by CBL in response to tumor target cells (Table 6). This deficiency of IFN-y production may be due to a dysfunction of accessory cells (43) or to the effect of suppressor cells or their mediators (30, 44). Since the fetus represents an allograft in the maternal environment, a decrease in the cytotoxic response of the feto-maternal unit may inhibit graft-vs-host reactions as well as facilitate the expansion and development of hematopoietic stem cells which are known to be regulated by NK cells (8). In view of the significant ability of IFN or IL2 to enhance the NK activity of neonatal lymphocytes as reported herein, these important biological modifier agents may have therapeutic potential in the treatment of severe neonatal viral infections. ACKNOWLEDGMENTS The authors express their sincere appreciation to Denise DuPrie for her expert secretarial help and to Katherine Albertson, Lynette McCurry, and Jean CiJik for their excellent technical assistance.

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.

Janeway, C. A., Arch. Dis. Child. 41, 358, 1966. Schaffer, A. J., and Avery, M. E., “Diseases of the Newborn,” p. 633. Saunders, Philadelphia, 1971. Dwyer, J. M., and Johnson, C., Cell. Zmmunol. 81, 81, 1983. Herberman, R. B. (Ed.), In “NK Cells and Other Natural Effector Cells” Academic Press, New York, 1984. Nab, M. P. N., Fernandes, A., Onoe, K., Day, N. K., and Good, R. A., Znt. J. Cancer 25,667, 1980. Nair, M. P. N., and Schwartz, S. A., J. Zmmunol. 129, 2511, 1982. Nair, M. P. N., and Schwartz, S. A., Cell. Zmmunol. 81, 45, 1982. Cudkowicx, G., and Hochman, P. S., Zmmunol. Rev. 44, 13, 1979. Kohl, S., Frazier, J. P., Pickering, L. K., and Loo, L. S., J. Pediatr. 98, 783, 1981. Antovelli, P., Stewart II, W., and DuPont, B., Clin. Zmmunol. Zmmunopathol. 19, 161, 1981. Kaplan, J., Shope, T. C., Bollinger, R. O., and Smith, J., J. Clin. Zmmunol. 2, 350, 1982. Hashimoto, G., Wright, P. F., and Katzon, D. T., Infect. Zmmun. 42, 214, 1983. Uksila, J., Lass& O., and Hirvonen, T., Clin. Exp. Zmmunol. 48, 649, 1982. Tarkkanen, J., and Saksela, E., Stand. J. Zmmunol. 15, 149, 1982. Boyum, A., &and. J. Clin. Lab. Invest. 21, Suppl. 97, 77, 1968. Nair, M. P. N., and Schwartz, S. A., J. Immunol. 126,2221, 1981. Nair, M. P. N., and Schwartz, S. A., J. Zmmunol. 132,2876, 1984. Kadish, A. S., Doyle, A. T., Steinhauer, E. M., and Ghossein, N. A., J. Zmmunof. 127, 18 17, 198 1.

ACTIVITY

OF CORD BLOOD

LYMPHOCYTES

171

19. Haliotis, T., Roder, J., Klein, M., Ortaldo, J., Fauci, A. S., and Herberman, R. B., .I. Exp. Med. 151, 1039, 1980. 20. Borden, E. C., and Leonhardt, P. H., J. Lab. Clin. Med. 89, 1036, 1911. 21. Fan-am, E., and Targan, S. R., .I Immunol. 130, 1252, 1983. 22. Santoli, D., Trinchieri, G., and Kaprowski, H., .I. Immunol. 121, 532, 1978. 23. Rosenberg, E. B., Herberman, R. B., Levine, P. H., Halterman, R. H., McCoy, J. L., and Wunderlich, J. R., Int. J. Cancer 9, 648, 1972. 24. Saksela, E., Timonen, T., and Cantell, K., &and. J. Immunol. 10, 251, 1979. 25. Abo, T., Cooper, M. D., and Balch, C. M., J. Exp. Med. 155, 321, 1982. 26. Frazier, J. P., Kohl, S., Pickering, L. K., and Loo, L. S., Pediatr. Rex 16, 558, 1982. 27. Rodriguez, M. A., Bankhurst, A. D., Ceuppens, J. L., and Williams, R. C., Jr., .I Clin. Invest. 68, 1577, 1981. 28. Maccario, R., Nespoli, L., Mingrab, G., Vitiello, A., Ugazio, A. G., and Burgio, G. R., .I Immunol. 130, 1129, 1984.

29. 30. 31. 32. 33. 34. 35. 36. 37. 38.

Witemeyer, S., Bankhurst, A. D., and Williams, R. C., Jr., Cell. Immunol. 30, 54, 1977. Olding, L. B., Murgita, R. A., and Wigzell, H., J. Immunol. 119, 1109, 1977. Timonen, T., and Saksela, E., J. Immunol. Methods 36,285, 1980. Kohl, S., Loo, L. S., and Gonik, B., J. Infect. Dis. 150, 14, 1984. Wright, S. C., Weitzen, M. L., Kahle, R., Granger, G. A., and Bonavida, B., J. Immunol. 130, 2479, 1983. Herberman, R. B., Ortaldo, J. R., and Bonnard, G. D., Nature (London) 277, 22 1, 1979. Lattime, E. C., Pecoraro, G. A., and Stutman, O., J. Exp. Med. 157, 1070, 1983. Perussia, B., and Trinchieri, G., J. Immunol. 126, 154, 1981. Bishop, G. A., and Schwartz, S. A., Clin. Immunol. Immunopathol. 25, 374, 1982. Ortaldo, J. R., Pestka, S., Slease, R. B., Rubinstein, M. I., and Herberman, R. B., &and. J. Immunol. 12, 365, 1980.

39. Pearlstein, K. T., Palladino, M. A., Welte, K., and Vilcek, J., Cell. Immunol. 80, 1, 1983. 40. Svedersky, L. P., Shepard, H. M., Spencer, S. A., Shalaby, M. R., and Palladino, M. A., J. Immunol. 133, 714,

1984.

4 I. Miyawaki, T., Seki, H., Kubo, M., and Taniguchi, N., J. Immunol. 123, 1092, 1979. 42. Bryson, Y. J., Winter, H. S., Gard, S. E., Fischer, T. J., and Stiehm, E. R., Cell. Immunol. 1980. 43. Epstein, L. B., Cline, M. J., and Merigan, T. C., J. Clin. Invest. 50, 744, 1971. 44. Hayward, A. R., and Lawton, A. R., J. Immunol. 119, 1213, 1977.

55,

191,